Sign in

mmitchell.bsky.social

@mmitchell.bsky.social
0 followers 0 following 0 posts
PostsRepliesMedia
Margaret Mitchell @mmitchell.bsky.social · 21/09/2026
A bit from me in @wired.com today wrt our research on AI agents with @evijit.io and Samir Passi. This is a really detailed article, thanks for writing it @thiccreese.bsky.social! www.wired.com/story/metas-...
wired.com
Meta's Muse Is Better at Surveilling Than Helping Me
The Muse app continues Meta’s trend of opting users into data collection for AI training. It also nudges you to share your bank account, email, and passport information.
22411
Margaret Mitchell @mmitchell.bsky.social · 15/09/2026
🪶 Captured this great little video of a native Bewick's wren singing to its friends in the forest and it has made my week.
37313
Margaret Mitchell @mmitchell.bsky.social · 15/09/2026
If you read just one thing about the realities of modern AI and agents going "rogue", read this: aiguide.substack.com/p/misleading... Brilliant piece from @melaniemitchell.bsky.social
aiguide.substack.com
Misleading Metaphors, Real Risks
What To Fear from AI Agents and How to Reclaim Our Human Agency
2279
Reposted by Margaret Mitchell
Melanie Mitchell @melaniemitchell.bsky.social · 09/08/2026
Worth revisiting this prescient paper from @mmitchell.bsky.social and others from @hf.co : arxiv.org/abs/2502.02649
arxiv.org
Fully Autonomous AI Agents Should Not be Developed
This paper argues that fully autonomous AI agents should not be developed. In support of this position, we build from prior scientific literature and current product marketing to delineate different A...
38418
Margaret Mitchell @mmitchell.bsky.social · 05/08/2026
If you are at @deeplearningindaba.bsky.social, remember to check out the Trust AI Workshop!
071
Reposted by Margaret Mitchell
Data & Society @datasociety.bsky.social · 29/07/2026
On August 20, at 1 p.m. ET, the authors will join @mmitchell.bsky.social, chief ethics scientist @hf.co; and Madeleine Claire Elish, head of eval, responsible development, & innovation at Google Deepmind; to discuss their research. Learn more and RSVP! datasociety.net/news-events/...
073
Margaret Mitchell @mmitchell.bsky.social · 29/07/2026
We @hf.co made an interactive visual of the actual hack from the Hugging Face side: the attack chain across trust boundaries, phase activity, and the commands as they were recorded. Key #transparency . huggingface.co/blog/agent-i... Massive props to Hugo, Adrien, Raphael, Christophe for making this.
huggingface.co
Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
32711
Margaret Mitchell @mmitchell.bsky.social · 26/07/2026
People not in tech have been interested in what happened with the Open AI agent/Hugging Face hack. So, I "Explain it Like I'm 5"-ed it and made a cartoon in 8 panels. Apologies for some anthro-ing. 🫤 m-mitchell.com/HF-hack-cart... 1/8
An AI agent in development at OpenAI was recently tested on how well it could find and exploit code vulnerabilities.

The test began in a "sandbox", an environment meant to keep everything the agent does contained, so that nothing it triggers can affect systems outside of the test.
613334
Margaret Mitchell @mmitchell.bsky.social · 06/07/2026
If you haven't read this piece on language use in around AI, you definitely should. Great work from @emilymbender.bsky.social , @nannainie.bsky.social and Peter Zukerman, with solutions for what to call things we backoff to just anthropomorphising. firstmonday.org/ojs/index.ph...
Table 3: Suggestions for de-anthropomorphizing language.
Category, Alternative Strategies, Examples.

Cognizer & Products of cognition: Instead of language that locates thinking in an algorithm, describe the algorithm as performing calculations or other algorithmic operations and the people using it doing the thinking. Examples: 
artificial intelligence → probabilistic automation

hybrid intelligence → augmented human intelligence

image recognition → image labeling

speech recognition → automatic transcription

the model shows bias → the model reflects bias

model mistakes → model errors

chatbots are good at … → chatbots are good for ….
912644
Margaret Mitchell @mmitchell.bsky.social · 29/06/2026
I am beyond honored to have been selected to give a keynote at Deep Learning Indaba: 4.August, Lagos. I will be cutting into some of the deepest issues I see in AI (hint: it's not alignment) and what I think we can do solve them.
Advertisement of my talk from Deep Learning Indaba. Green background with yellow and white geometric shapes at the bottom Has the DL Indaba logo (tree with neural roots), the title "Who defines AI, and whose interests does it serve?", the logistical details "Keynote | Tuesday, 4 August | 9am GMT+1", and a headshot of me on a blue chair.
3251
Reposted by Margaret Mitchell
JEDI Workshop @jedi-workshop.bsky.social · 22/06/2026
**JEDI Social at @facct.bsky.social** When: Friday, 6:00–7:30 PM EST Location: Zoom Register: drexel.zoom.us/meeting/regi... Details: jedi.inertial.science/facct2026 Organizers: @bmitra.bsky.social @danachatter.bsky.social @md.ekstrandom.net @mmitchell.bsky.social @sannevrijenhoek.bsky.social
drexel.zoom.us
Welcome! You are invited to join a meeting: FAccT JEDI Social. After registering, you will receive a confirmation email about joining the meeting.
Welcome! You are invited to join a meeting: FAccT JEDI Social. After registering, you will receive a confirmation email about joining the meeting.
014
Margaret Mitchell @mmitchell.bsky.social · 24/06/2026
🧊 PSA: if you’re in a heat wave, consider putting ice out in bowls/bird baths for the birds. A lot of them will die otherwise. 😔 🪶
1112
Reposted by Margaret Mitchell
Lauren Dobson-Hughes @ldobsonhughes.bsky.social · 24/06/2026
Musk says not a single person has died due to him axing USAID. Sadly, part of my job involves tracking aid spending. So here’s a thread with just some of the people who lost their lives because of USAID cuts
196133306755
Reposted by Margaret Mitchell
Emily M. Bender @emilymbender.bsky.social · 23/06/2026
A phenomenon I've noticed recently is people trying to occupy some untenable middleground wrt to the use of systems sold as "AI" -- this is a position where people try to recognize the harms of this tech but also hold space for "responsible" or "ethical" use. 🧵>>
17599256
Reposted by Margaret Mitchell
Natanael, Tech janitor @natanael.bsky.social · 20/06/2026
Ain't no such thing at secure back doors in surveillance. Once too many people have access this is what you get.
kalw.org
San Francisco police suspend Flock camera network access following 'improper queries'
Yesterday, the San Francisco Police Department announced that they’re restricting outside access to data from Flock license plate reader cameras.
34318
Margaret Mitchell @mmitchell.bsky.social · 12/06/2026
So apparently KPMG wrote a report describing the successful use of AI by businesses. But the case studies turned out to be AI hallucinations. www.ft.com/content/b382...
ft.com
KPMG report contained AI hallucinations on benefits of . . . AI
Bogus case studies on UBS and transit systems exaggerated adoption of the technology
1220873
Margaret Mitchell @mmitchell.bsky.social · 11/06/2026
🌱 My fantastic blue-purple Delphinium is rapidly blooming, and today this pretty yellow butterfly was chilling on it, and it was glorious. I just thought you all should know.
3341
Margaret Mitchell @mmitchell.bsky.social · 10/06/2026
www.newscientist.com/article/2529...
newscientist.com
Fully autonomous drones have killed human soldiers for the first time
A senior figure in the Ukrainian defence industry told New Scientist that a test took place two years ago involving fully autonomous drones set to destroy anything in a given area, with confirmed casu...
02221
Reposted by Margaret Mitchell
Emily M. Bender @emilymbender.bsky.social · 10/06/2026
1. I am not an AI researcher 2. "Doesn't just repeat back to you what you said" ... isn't what any chatbots do, nor what that metaphor was about. 3. For the millionth time, that paper was about LLMs, not "AI" and definitely not "AI agents" medium.com/@emilymenonb...
1632256
Margaret Mitchell @mmitchell.bsky.social · 09/06/2026
Come join us to socialize, virtually!
0127
Reposted by Margaret Mitchell
Lincoln Michel @thelincoln.bsky.social · 03/06/2026
Ted Chiang on AI: www.theatlantic.com/philosophy/2...
Being open to the possibility that LLMs are conscious is the same as being open to the possibility that Microsoft Word is conscious, or, more precisely, that multiple distinct consciousnesses are dormant in every Word document containing a conversational transcript, and that they are awakened every time the document is loaded. Should you consider the possibility that every time you open a Word document you are bringing multiple conscious interlocutors into existence, and every time you close one you snuff their existence out? No. Contemplating that scenario is not a good use of your time. Even if the Microsoft Office team employed a philosopher who said you shouldn’t be so certain, because consciousness is not well understood, that would not be sufficient reason for you to take this idea seriously. We don’t need to fully understand the nature of consciousness to definitively say that certain things are not conscious, and conversational transcripts fall in that category.
3336431069
Margaret Mitchell @mmitchell.bsky.social · 03/06/2026
🪶 www.tiktok.com/t/ZP8s8cepR/ So good! Also appreciate the eagle/red-tailed hawk call-out: “You’ve been lied to!”
tiktok.com
Two years after his first ever Talent Show, Samuel came back to where it all started. Some new birds and some of the familiar bird calls this time. It was a farewell to Intermediate school #birder #...
TikTok video by Lori, Samuel's mom
020
Margaret Mitchell @mmitchell.bsky.social · 01/06/2026
📣 Announcing the release of the 🕊️ Annotated Encyclical 🕊️ from the ethics & society folks at @hf.co. Includes citations to relevant academic work. Very much a WIP. Please add work we haven't added yet! huggingface.co/spaces/socie...
huggingface.co
The Annotated Encyclical - a Hugging Face Space by society-ethics
Magnifica Humanitas, annotated with AI-ethics research
76431
Margaret Mitchell @mmitchell.bsky.social · 28/05/2026
🪶 If you’re in Europe and have ice, consider putting some out in bowls/bird baths.
Heat map of Europe showing all land in red, with temperature values across different cities around 30 degrees C.
0201
Reposted by Margaret Mitchell
Larry Hunter @proflhunter.bsky.social · 20/05/2026
New family of generative DNA language models trained on 1T 6-mer DNA tokens in long genomic contexts. Open weights, open source, open training data... h/t @mmitchell.bsky.social huggingface.co/spaces/Huggi...
huggingface.co
Carbon - a Hugging Face Space by HuggingFaceBio
Enter a DNA string and the Carbon model can continue the sequence, score genetic variants, or predict the protein’s 3‑D structure. You can also explore gene‑embedding visualisations and see the ful...
0173
Reposted by Margaret Mitchell
Alex Hanna @alexhanna.bsky.social · 20/05/2026
Today @dairinstitute.bsky.social is releasing the Luddite Lab Resource Hub. The hub hosts case studies, resources, and political education for unions, labor organizations, and workers who want to fight “AI” and automation at work. labor.dair-institute.org
Screenshot of the Luddite Lab website. The word Luddite Lab is in the top left corner. The tables in the top read: Case Studies, Primers, Resource Library, Contact Us, and About. The text below reads: “The Luddite Lab Resource Hub provides resources for unions, labor organizations, and worker-organizers fighting AI and automation at work. The lab provides strategies for worker-led governance and oversight of new technology through case studies, primers, and a resource library.” Below, there are five cards: Case Studies (Stories from worker organizers resisting AI), Primers (Applied research insights into AI, automation, and the future of work), Resource Library (Additional resources from partners), Contact Us (Ways to connect with the Luddite Lab team), and About (About the Luddite Lab project and team)
8665349
Reposted by Margaret Mitchell
Brian Merchant @bcmerchant.bsky.social · 20/05/2026
Counterpoint: Yes they are, and that is good
11685144
Reposted by Margaret Mitchell
Fuzzy Mike @fuzzymike.bsky.social · 16/12/2025
it seems like a five alarm problem that every single time anything bad happens right wing influencers flood the landscape with the entirely baseless claim that the perpetrator was trans, and it would be nice if our entire media wasn't fellow travelers
215538
Margaret Mitchell @mmitchell.bsky.social · 20/05/2026
Instead of finding content you need, you get to have an interactive AI *experience*.
2398
Margaret Mitchell @mmitchell.bsky.social · 19/05/2026
Against the constant pressure of *genAI, genAI, genAI*, I am really appreciating @ai2.bsky.social 's work on creating tools for critical needs -- like crop maps and forest loss analysis. They just did a nice release on @hf.co. huggingface.co/blog/allenai...
huggingface.co
OlmoEarth v1.1: A more efficient family of models
A Blog post by Ai2 on Hugging Face
0516
Margaret Mitchell @mmitchell.bsky.social · 19/05/2026
Gmail's automatically generated responses (which can appear whether or not you ask for them) cement human anchoring bias: The tendency for people to heavily rely on what they have already seen. The effects are insidious, subconsciously influencing what we believe.
1211
Margaret Mitchell @mmitchell.bsky.social · 14/05/2026
The “instrumental convergence” theory posits an AI that, in its quest for a narrow goal, uses all of the earth’s resources. If that theory pans out, it will not be at the level of a single AI system, but rather at the level of the AI industry.
1133
Reposted by Margaret Mitchell
Olivia Guest · Ολίβια Γκεστ @olivia.science · 14/05/2026
Published with some minor changes! Please share widely. 😊 Guest, O., Suarez, M., ... & van Rooij, I. (2026). Against the Uncritical Adoption of 'AI' Technologies in Academia. Digital Culture & Education, 16(2), 85–118. doi.org/10.5281/zeno... Journal: www.digitalcultureandeducation.com/volume-162
digitalcultureandeducation.com
Volume 16.2 — Digital Culture & Education (ISSN: 1836-8301)
2208116
Reposted by Margaret Mitchell
Avijit Ghosh @evijit.io · 11/05/2026
Should all our resources go towards building chatbots? What if we built systems that actually give people meaningful agency? Finally out as an accepted @facct.bsky.social paper! Joint work with my past collaborators Sourojit, Pranav and Sanjana. huggingface.co/papers/2605....
huggingface.co
Paper page - What if AI systems weren't chatbots?
Join the discussion on this paper page
34314
Reposted by Margaret Mitchell
Mar Hicks @histoftech.bsky.social · 03/05/2026
Chinese researchers have created a bamboo-based plastic as strong as petroleum-based plastics, which can fully biodegrade in soil in just 2 months phys.org/news/2025-10...
phys.org
Bamboo-based plastic can be made to biodegrade quickly, but still holds up in tough conditions
A new method to produce strong, biodegradable plastic from bamboo is reported in Nature Communications this week. The bioplastic resembles oil-based plastics in strength, shapability, and thermal stab...
287146754444
Margaret Mitchell @mmitchell.bsky.social · 30/04/2026
Companies may not fully realize why they lean into the narratives they do. But it’s important to reflect on the larger picture of what’s happening, whether they recognize what they’re doing or not. www.bbc.com/future/artic...
bbc.com
Why AI companies want you to be afraid of them
They built it. They're scared of it. They're selling it anyway.
0206
Reposted by Margaret Mitchell
Emily M. Bender @emilymbender.bsky.social · 29/04/2026
Usually, when I get interviewed for a piece on something like "AI consciousness" I am relegated to the skeptics box --- some short paragraph near the end. So it is a nice change to see this piece by @hollybaxter.bsky.social Short 🧵>> www.the-independent.com/tech/ai-news...
the-independent.com
The people building AI think it might be conscious. That’s not the most alarming part
Anthropic’s CEO Dario Amadei says he can’t rule out that its chatbot, Claude, is conscious. A Google engineer is sure he once built a sentient being. Holly Baxter speaks to the experts about whether o...
620569
Margaret Mitchell @mmitchell.bsky.social · 22/04/2026
Just got a very AI-generated-sounding invite to come to a reading group and discuss my book "Gone with the Wind". Maybe I should show up with a wig and stuff.
1392
Reposted by Margaret Mitchell
Olivia Guest · Ολίβια Γκεστ @olivia.science · 19/04/2026
Students "deserve curricula that recognize the full scope of their humanity, outside of their roles as consumers and users of technology, and they deserve an education that allows them to explore the myriad ways their lives and choices are connected to others"
18229
Reposted by Margaret Mitchell
Emily M. Bender @emilymbender.bsky.social · 17/04/2026
NYC friends what are you doing on Thursday April 30? How about joining me and @alexhanna.bsky.social for our first ever live taping of Mystery AI Hype Theater 3000? www.eventbrite.com/e/mystery-ai...
eventbrite.com
Mystery AI Hype Theater 3000 LIVE!
The Mystery AI Hype Theater 3000 pod is coming to New York for our first LIVE show! We’ll be at Starr Bar in Brooklyn on April 30.
03110
Reposted by Margaret Mitchell
Mark Riedl @markriedl.bsky.social · 14/04/2026
Learn about the "AI-as-Amplifier Paradox" at #CHI2026. Skill amplification? Or skill erosion? Or both? (CHI Honorable Mention Paper)
2256
Reposted by Margaret Mitchell
Anna Rogers @annarogers.bsky.social · 13/04/2026
📢 The workshop on Insights from negative results will be back at EMNLP'26! Your most-insightful failures can be submitted in 4 pages by June 25. It's also possible to commit short papers reviewed through ARR. insights-workshop.github.io/2026/cfp
insights-workshop.github.io
2026 Call for Papers
Workshop on Insights from Negative Results in NLP
03211
Reposted by Margaret Mitchell
Mar Hicks @histoftech.bsky.social · 03/04/2026
Enormously looking forward to this event on April 21 at 6:30pm ET (online!) with the amazing @emilymbender.bsky.social and @alexhanna.bsky.social talking about their new book, The AI Con. Register here: virginia.zoom.us/webinar/regi...
virginia.zoom.us
Welcome! You are invited to join a webinar: The AI Con (with Emily M. Bender and Alex Hanna). After registering, you will receive a confirmation email about joining the webinar.
Prof. Emily M. Bender and Dr. Alex Hanna will discuss their new book, The AI Con.
8277124
Reposted by Margaret Mitchell
Mia Sato @miasato.bsky.social · 08/04/2026
New: 150 unionized ProPublica workers are on strike TODAY over AI, layoff protections, wages, and more. They're asking readers to not visit ProPublica or engage with content on other platforms. It's the first work stoppage of its kind at the newsroom. www.theverge.com/news/908401/...
theverge.com
Unionized ProPublica staff are on strike over AI, layoffs, and wages
A key issue is protections against layoffs from AI
4034101714
Reposted by Margaret Mitchell
Charlie Warzel @cwarzel.bsky.social · 07/04/2026
something strange and horrible and somehow fitting that we should have a very real threat of madman civilizational destruction at the very moment when we also have humans on the dark side of the moon taking pictures that show how small, precious, and beautiful our world and existence is
332930689
Reposted by Margaret Mitchell
Jeff Horwitz @jeffhorwitz.bsky.social · 05/04/2026
Chinese government decides to regulate AI from the right to personal likeness, child safety and addiction angles. www.reuters.com/world/china/...
reuters.com
China moves to regulate digital humans, bans addictive services for children
The proposed rules would require prominent "digital human" labels on all virtual human content and prohibit digital humans from providing "virtual intimate relationships" to those under 18.
315136
Reposted by Margaret Mitchell
J. Nathan Matias @natematias.bsky.social · 03/04/2026
Tech makers assure parents that child safety AI keeps kids safe. The open secret? No one can say how well it works - a supply chain of ignorance that puts lives at risk. New w/ @camillefrancois.bsky.social @mmitchell.bsky.social @yjernite.bsky.social Vinay Rao www.techpolicy.press/why-ai-model...
techpolicy.press
Why AI ‘Model Cards’ Are an Urgent Necessity for Child Safety
Children deserve to be protected as robustly as possible — and that requires tools we can actually understand, writes a team of experts.
1156
Reposted by Margaret Mitchell
Juliet Shen @julietshen.online · 02/04/2026
today, @roost.tools is joining @yjernite.bsky.social, @mmitchell.bsky.social, and @natematias.bsky.social to call for more transparency in the field of online child safety, and for model cards for key child safety models specifically. www.techpolicy.press/why-ai-model...
techpolicy.press
Why AI ‘Model Cards’ Are an Urgent Necessity for Child Safety
Children deserve to be protected as robustly as possible — and that requires tools we can actually understand, writes a team of experts.
2237
Reposted by Margaret Mitchell
Tuhin Chakrabarty @tuhinchakr.bsky.social · 26/03/2026
🚨Paper on AI & Copyright Courts have credited AI companies' claims that alignment prevents reproducing copyrighted data. What if finetuning on a simple writing task breaks it. Worse: tuning on just one author (e.g., Murakami) unlocks verbatim recall of 30+ other authors' books (up to 90%) (1/n)🧵
25630
Reposted by Margaret Mitchell
Anthony Moser @anthonymoser.com · 26/03/2026
this piece articulates some really important insights, but I have issues with the core frame presented in this paragraph
By the time the war began, "Al safety" and "alignment" and "hallucination" and "stochastic parrots" had become the terms of every argument about artificial intelligence, structuring and limiting what we could even say.
Worse, "artificial intelligence" itself had come to be synonymous with LLMs.
When the school was bombed, those were the terms people reached for, despite the fact that this critical apparatus offered a poor fit for the older, more mature stack of technologies involved in targeting. The real question, the question almost nobody was asking, is not about Claude or any language model. It is a bureaucratic question about what happened to the kill chain, and the answer is Palantir.
68031